ESSENTIALS
LE ARN -.
CODEf^
I WITH t-l
PROGRAM WITH THE WORLD'S
MOST POPULAR LANGUAGE
YOUR Raspberry Vi
Written by Simon Long
raspberrypi.org/magpi
SUBSCRIBE
TODAY AND
RECEIVE A
FREE
PI ZERO W
Subscribe in print for 12
months today and receive:
Other benefits:
(before stores)
The office
Bsassssw
gpM'g *
,d new RaP __-
• A free Pi Zero W (the latest model)
• Free Pi Zero W case with 3 covers
• Free Camera Module connector
• Free USB and HDMI converter cables
• Save up to 23% on the price
• Free delivery to your door
• Exclusive Pi offers & discounts
• Get every issue first
PLUS
AN OFFICIAL
PI ZERO CASE
WITH 3 COVERS
AND FREE CAMERA MODULE
CONNECTOR AND USB / HDMI
CONVERTER CABLES
Pricing:
Get six issues:
£30 (UK)
£45 (EU)
$69 (USA)
£30 (Rest of World)
Get three issues:
£12.99 (uk) (Direct Debit) I $37.30 (us) (quarterly)
How to subscribe:
# magpi.cc/Subs-2 (UK / ROW) # imsnews.com/magpi (USA)
# Call +44(0)1202 586848 (UK/ROW) # Call 800 428 3003 (USA)
Search ‘The MagPi’ . getiton
on your app store: Google Play
WELCOME \
TO LEARN TO \
CODE WITH C \
a "\ he C programming language was
| B invented in the early 1970s, and
since then has become one of the
most popular and widely used general-
purpose languages. It’s used by a wide range
of programmers, from amateurs working
on simple projects at home, to industry
professionals who write in C for a living. It’s
been used to program everything from the tiny
microcontrollers used in watches and toasters ^
up to huge software systems - most of Linux ^
(and Raspbian itself) is written in it. It can
give you control over the smallest details of
how a processor operates, but is still simple to
learn and read. This series is an introduction
to programming in C for absolute beginners;
you don’t need any previous programming
experience, and a Raspberry Pi running
Raspbian is all you need to get started.
Simon Long
FIND US ONLINE raspberrypi.org/magpi GET IN TOUCH magpi@raspberrypi.org
MatJPi
B AppStore
EDITORIAL
Managing Editor: Russell Barnes
russell@raspberrypi.org
Author: Simon Long
Sub Editors: Lorna Lynch and Laura Clay
DISTRIBUTION
Seymour Distribution Ltd
2 East Poultry Ave. London
EC1A9PT | +44(0)2074294000
DESIGN
Critical Media: criticalmedia.co.uk
Head of Design: Dougal Matthews
Designers: Lee Allen, Mike Kay
THE MAGPI SUBSCRIPTIONS
Select Publisher Services Ltd
PO Box 6337. Bournemouth
BH19EH | +44(0)1202586848
magpi.cc/Subsi
Qr recycle
In print, this product is made using paper
sourced from sustainable forests and
the printer operates an environmental
management system which has been
assessed as conforming to ISO 14001.
This book is published by Raspberry Pi (Trading) Ltd., 30 Station Road, Cambridge, CBi 2JH. The
publisher, editor and contributors accept no responsibility in respect of any omissions or errors
relating to goods, products or services referred to or advertised in this product. Except where
otherwise noted, content in this magazine is licensed under a Creative Commons Attribution-
NonCommercial-ShareAlike 3.0 Unported (CC BY-NC-SA3.0). ISBN: 978-1-912047-03-1
06
GETTING STARTED
What's so great about C?
11 [
VARIABLES & ARITHMETIC
Create variables and do maths
16 [CHAI
CONDITIONS
& COMPARISONS
Control the flow of your C programs
22 [CHAPTER
MORE ADVANCED
FLOW CONTROL
For loops and case statements
29
POINTERS
Variables have addresses too
CONTENTS
35
FUNCTIONS
Split your code into bite-sized chunks
42
ARRAYS & STRINGS
Handle lists of values, and letters
48
THE STRING LIBRARY
Simplify common operations on strings
55 [ I
USER INPUT
Reading and interpreting user input
62
FILE INPUT AND OUTPUT
Learn to read from and write to files
68
MORE ABOUT TYPES
AND VARIABLES
Type definitions, enumerations & more
74
HEADER FILES AND
THE PREPROCESSOR
Splitting code up into multiple files
81 [C
QUICK REFERENCE
Cheat sheets and code examples
Simon Long is a
software engineer
at Raspberry Pi,
with a particular
interest in user
interface design.
He first started
writing C in the
early 90s, and
since then he's
written software
for a wide variety
of electronic
products, from
mobile phones
to medical
equipment.
Nowadays, he
spends most of
his time fiddling
with the Raspbian
desktop trying
to make it better,
and when he's
not doing that,
he's usually
listening to 80s
progressive rock
and solving those
really difficult
crosswords
without any
black squares.
[ Contents ] MatfPi
5
The jT\9
ESSENTIALS
CHAPTER ONE
GETTING
STARTED
C is one of the most widely used programming
languages - learn how to use it to program the Pi!
[ LEARN TO CODE WITH C ]
I © a ■ # © !■ pi@raspberry: ~ | »|phello.c - /home/pi - G...
[CHOOSE
YOUR EDITOR ]
You can use
whatever
editor you Like
to enter code,
as Long as it
saves it as
plain text. The
Geany editor
included in
Raspbian is a
good choice,
but you can
also use
Leafpad, nano,
or any others
that you prefer.
What's so great about C?
C is a very powerful language - there’s not much you can’t use it for -
but it’s fairly simple. The language itself only has 20 or so keywords,
but there’s a huge library of additional functions that you can call in
when you need them. In this series, we’re going to concentrate on
learning about the keywords, with a few of the more useful library
functions thrown in for good measure.
Many of the languages that you may have seen, such as Python,
are what are called interpreted languages. This means that the code you
write is run directly: each line of code is read in and interpreted as you
run it. C is different: it’s a compiled language. This means that the code
you write, known as the source code, is never run directly. The source
code is passed through a program called a compiler, which converts it
into a machine-readable version called an executable or a binary; you
then run the resulting executable.
[ Getting Started ] 7
[ WHITESPACE
DOESN'T
MATTER! ]
Unlike Python,
whitespace
has no
significance
in C - you can
put spaces,
tabs, and new
lines anywhere
you like in a
C program
to make it
readable. i
This may seem complex, but it has a few big advantages. First,
it means that you don’t need to have a copy of C itself on every
computer you want to run your program on; once compiled,
the executable is stand-alone and self-contained. Second, the
compilation process will find a lot of errors before you even run the
program (but it won’t usually find all of them). Most importantly, the
compilation process means that the time-consuming translation of
human-readable code into machine-readable instructions has already
happened, which means that compiled code generally runs many
times faster than interpreted code would.
Hello world - your first C program
With all that out of the way - which has hopefully made you think
that C might be worth learning - let’s have a look at the first program
everyone writes in any language, the one that prints ‘Hello World’ on
the screen. Incidentally, the tradition of writing a Hello World program
was first introduced with the original documentation describing C
itself. Just think: no C, no Hello World...
ttinclude <stdio.h>
void main (void)
{
/* A print statement */
printf ("Hello world!\n");
}
Hopefully not too frightening! Let’s look at it line by line.
#include <stdio.h>
This is known as a hash-include. As mentioned above, the C
language has a large library of functions that can be included, and
we need to use one of them in this program: the formatted print
command printf. This is part of the standard input-output library,
or stdio for short. So what this line does is to warn the compiler that
the program needs the stdio library to be included as part of the
compile process.
8
IMat/Pi [ Chapter One ]
void main (void)
C is a function-based language; every program is made up
of a number of functions.
Each function takes zero or more arguments, and returns a single
value. A function definition consists of a specification of what the
function returns (in this case, a void), a function name (in this
case, main), and a list of arguments enclosed in round brackets
(again, a void).
Every C program has to include a function called main; when you run
the compiled program, the main function is the first thing that executes.
The word void is called a type specifier; a void is a special type which
means ‘no value required’. We’ll look more at types in the next chapter.
So this line defines the main function for this program; it states that
the main function takes no arguments, and returns no value.
The code which makes up the function itself is enclosed between
the two curly brackets {} that follow the function definition.
/* A print statement */
First, we have a comment telling us what’s going on. Comments in
C start with the symbol /*, and end with */ - anything between those
two symbols is ignored by the compiler.
The code itself is just one line:
printf ("Hello world!\n");
This is a call to the printf (‘print formatted’) function from the
stdio library. In this case, it takes a single argument, which is a text
string enclosed within double quotes. As mentioned above, function
arguments are enclosed in round brackets.
Note that the line ends with a semicolon. All statements in C must
finish with a semicolon; this tells the compiler that this is the end
of a statement. One of the most common beginner mistakes in C
is to forget a semicolon somewhere!
What about the string itself? The Hello World ! bit is
straightforward enough, but what about that \n at the end? Remember
this function is called ‘print formatted’? Well, the \n is a bit of
[CHECK YOUR
BRACKETS]
Unlike
whitespace,
punctuation is
very important
in C - make
sure you don't
use a curly
bracket where
a round one
is needed, or
vice versa.
[ Getting Started ] ‘MatfPi
9
ESSENTIALS
Right You interact
with both the
C compiler and
your compiled C
programs from the
command line; you
can either do this in
a terminal window
in the desktop, or
by booting your
Pi straight to the
command line
File Edit Tabs Help
piQraspberry $ gcc -o myprog he
piQraspberry :- $ ./myprog
Hello world!
sllo.c
pigraspberry $ |
formatting; it’s the symbol for a newline character. So this line will
print the string ‘Hello World!’, followed by a new line.
Compiling your program
Let’s compile and run this. Raspbian includes a C compiler called gcc,
so there’s nothing to install; just start up Raspbian on your Pi and
you’re ready to go. Use your favourite text editor to create a file called
hello, c, copy the program above into it, and save it. Then, from
a terminal, go into the directory where you saved hello. c and enter:
gcc -o myprog hello.c
[ RUNNING
YOUR
PROGRAM ]
You need to
tell Linux that
the program
you want
to run is in
the current
directory, so
don't forget
the . / before
This calls the gcc C compiler with the option -o myprog, which tells
it to create an executable output file called myprog, and to use hello. c
as the input source code.
If you entered your C code correctly (did you make sure the semicolon
was there?), this should take a second or so and then return you to
the command line. There should now be a file in the current directory
called myprog - try running it by typing:
./myprog
Et voila! You should now have...
Hello World!
...written in the terminal.
That’s your first C program written, compiled, and run. In the next
chapter, we’ll start using C for something a bit more useful...
10 I MatfPi [ Chapter One ]
ESSENTIALS
CHAPTER TWO
VARIABLES &
ARITHMETIC
Doing some real work in C: creating
variables and performing mathematical
operations on them
?nu ] © L_ ■ ♦ Q • . • math c - /home/pi - G... |J B pi@raspberry: ~
9
'Jp math.c - /home/pi - Geany
File Edit Search View Document Project Build Tools Help
C? v Q v & x * - 4 $
<J Symbols I math.c
eft Functions
<P main [3|
hFPl
□□□
File Edit Tabs Help
^include <stdio.h>
void main (void)
pi®raspberry:- $ . /myprog
The result of adding 2 and 3 is 5
pi@raspberry:~ $ |
printf ("The result of adding %d and sd is %d\n‘> a, b, c);
These are variable declarations:
in C, a variable must be
declared before you use it
C can print the results of
calculations to the terminal
in formats you choose
_IL
[ MULTIPLE
DECLARATIONS ]
You can
declare
multiple
variables of
the same type
in one line,
separated by
commas. For
the example
here, instead of
three separate
int declarations,
you could type
int a, b =
3, c;
on one line.
O n some languages, you can create variables as you go along
and put whatever data you want into them. C isn’t like that: to
use a variable in C, you need to have created it first, and at the
time you create it, you have to set what type of value it’s going to store.
By doing this, a block of memory of the correct size can be allocated by
the compiler to hold the variable. This process of creating a variable
is known as declaration.
Integers
There are several fundamental data types in C, but we’ll start by
looking at one of the most commonly used: the int type, used
to store an integer value.
#include <stdio.h>
void main (void)
{
int a;
int b = 3;
int c;
printf ("The sum of adding %d and %d is %d\n"j a, b , c);
The top three lines inside the main function here are declarations.
They tell the compiler that we would like to use variables called a, b
and c respectively, and that each one is of type int, i.e. an integer.
In the second line, we see an example of an initialisation at the same
time as a declaration: this stores an initial value of 3 in the variable b.
Note that the values of a and c at this point are undefined; you might
assume that a variable which hasn’t had a value stored in it is always 0,
but that isn’t the case in C. Before reading the value from a variable or
using it in a calculation, you must store a value in it; reading a variable
before initialising it is a common error in C.
The next two lines do some actual work with the variables
we have declared.
a = 2;
[ ARITHMETIC
SHORTHAND]
C allows
shortcuts for
some common
operations;
for example,
instead
of typing
a = a + 1,
you can just
enter a++.
Or for
a = a * B,
you can enter
a *= 3
This stores a value of 2 in the variable a, which will now have this value
until it’s changed. The reason a is called a variable is that it can vary: you
can change its value as often as you like, but only to another integer. The
value of a variable can change, but its type is fixed when it is declared.
c = a + b;
This line adds a to b, and stores the result in c.
printf ("The sum of adding %d and %d is %d\n", a, b, c);
This is another use of the formatted print function we saw in the previous
chapter. Note the three %d symbols inside the string: these are format
specifiers, and they are how you output numbers in C. When the printf
function is executed, each %d is replaced by a decimal representation (d
for decimal integer) of the variable in the corresponding position in the
list after the string. So the first %d will be replaced by the value of a, the
second with the value of b, and the third with the value of c .
Compile the program above and then run it. You should see this:
The sum of adding 2 and 3 is 5
1
I
0
...in the terminal.
[ Variables & Arithmetic ] IWacfPi
13
ESSENTIALS
Floating-point numbers
So we can add two integers together; what else can we do? One thing
we might want to do is to use floating-point numbers: numbers with a
decimal point. These have a different type, called float. Try changing
the code above so instead of:
This tells the compiler that a is now a floating-point value, rather
than an integer. Compile and run your program. What happens?
Oops! That doesn’t look right, does it? What has happened is that,
while the maths is still all correct, the printf statement is now wrong;
you’re telling it to print a, which is a floating-point value, as a decimal
integer. To fix that, change the first %d in the printf function to %f ,
which is the format specifier for a floating-point number, like this:
[ DECIMAL
PLACES ]
You can set
the number of
decimal places
to display for a
floating-point
type-specifier
in printf
by putting
a decimal
point and
the number
of places
between the %
and the f - so
%. 3f will show
a float value
with three
digits after the
decimal point.
Below Don't forget to use %f instead of %d as the print specifier
when changing the int values to float values in the example
14 I MatfPi [ Chapter Two ]
printf ("The sum of adding %f and %d is %d\n
That should produce something a lot more sensible when you run it.
This is an important lesson about C: it will do exactly what you tell it
to, even if it makes no sense. You told it to show you a floating-point
number as if it were a decimal integer, and the compiler assumed that
was what you wanted, even though the result was nonsense.
When you’re working with variables, always keep track of what
values you’re putting in what types, as it’s easy to introduce errors
by assuming a variable is of one type when it’s actually another. One
common error is to put the results of a calculation on floating-point
values into an integer.
Try this: make b a float as well (not forgetting to change its format
specifier in the printf), but leave c as an int, and set the two floats
to values with decimal points, like this:
[ REMEMBER
PRECEDENCE ]
C obeys the
common rules
for operator
precedence -
so a = a + 2
* 3 evaluates
the multiply
first and then
adds the
result, 6, to a.
You can use
round brackets
to change
precedence -
a = (a + 2)
* 3
gives 3a * 6.
float a;
float b = 3.641;
int c;
a = 2.897;
c = a + b;
printf ("The sum of adding %f and %f is %d\n", a, b, c);
You’ll see a result like:
The sum of adding 2.897000 to 3.641000 is 6
6? That’s not right! But it’s exactly what you’ve asked for. What the
compiler did was to add the two floating-point values together, and
got the answer 6.538, but you then told the compiler to put that into c,
an integer variable. So the compiler just threw away everything after
the decimal point! If you change c to a float, and change the final %d
to %f, you’ll find it gives the correct answer.
That gives you some idea about how C handles numbers, and how
you can use it for arithmetic; in the next chapter, we’ll look at how
to use the results of calculations to make decisions.
[ Variables & Arithmetic ] IWacfPi
15
I MadPi
ESSENTIALS (7
CHAPTER THREE
CONDITIONS &
COMPARISONS
Branches and loops: controlling
the flow of your C program
[ LEARN TO CODE WITH C ]
GJ
Wastebasket
'•IP test.c - /home/pi - Geany
File Edit Search View Document Project Build Tools Help
C9 v B v
<1 Symbols
(P Functions
($> main [3]
14 «
IB
4 a
a_
test.c x |
1 #include <stdio.h>
2
3 void main (void)
int a = 0 ;
if (a = 0 )
{
printf (“a is equal to 0\n‘);
else
{
printf ('a is not equal to 0\n‘);
1
Status
A simple if-else structure
in C, making a decision
based on the value of 'a'
15:31:02: File /home/pi/math.c saved.
15:31:26: File /home/pi/math, c saved.
lome/pi/math.c saved,
lome/pi/test.c saved.
| 3 INS TAB mode: Unix (LF) encoding: UTF-8 filetype:C
□BE
— pi@raspb.erry: ~
File Edit Tabs Help
piQraspberry $ gcc -o myprog test.c
piQraspberry: $ ./myprog
a is equal to 0
pigraspberry:- $ |
When you type a closing
bracket in Geany, the
corresponding opening
bracket is highlighted to
help avoid errors
[CURLY
BRACKETS]
Curly brackets
are used
to group
together a set
of statements
which always
execute
together. If
your Loop or
if statement
only needs
to execute
one single
statement,
you can leave
out the curly
brackets after
the test, but
this can make
the code's
purpose less
obvious to
a human!
O ne of the fundamentals of any programming language is the
ability to make conditional operations - to change the program’s
flow depending on the result of a test - and C allows you to do
this. In this chapter, we’ 11 look at how you test conditions within your C
programs, and how you use the results to determine what happens next.
In C, the mechanism for controlling flow based on testing a condition
is the if-else statement. Here’s a simple example:
#include <stdio.h>
void main (void)
{
int a = 0;
if (a == 0)
{
printf ("a is equal to 0\n");
}
else
{
printf ("a is not equal to 0\n");
}
>
[ Conditions & Comparisons ]
[ ELSE-IF ]
You can have
multiple else
statements
in one test.
Instead of
one simple
else for one
alternative, use
else if ()
with a new
test for each
alternative you
want. We'll
look more at
this in the next
chapter.
Here, the keyword if is followed by a test enclosed in round brackets,
in this case (a == 0). If the test evaluates as true, the operations
enclosed by the curly brackets after the test are executed.
This example also shows the use of an else clause. At the end of the
curly brackets around the operations which you want to execute if the
test is true, there’s an else followed by another set of curly brackets;
these contain the operations you want to execute if the original test
evaluated as false.
Try compiling the code above, and change the value with which
a is initialised to make sure it does what you expect.
= or ==
That’s all fine, but what’s this a == 0 all about? Surely if we want to
know whether a is equal to 0, we just put a = 0. Why the two equals
signs? Well, try replacing the double equals sign with a single equals
and see what happens.
This is a very important aspect of C syntax, and a common source of
bugs. The equals sign is used for two different things: one is to assign
a value to a variable, whereas the other is to test whether a variable is
equal to a value. A single equals sign (=) assigns a variable; a double
equals sign (==) tests a variable.
So the statement...
if (a == 0)
.. .tests to see if a is equal to o. If it is, then the test evaluates as true,
and the code immediately after the if is executed.
But the statement...
if (a = 0)
.. .doesn’t compare a against 0 at all: it just sets a to o. So how does the
compiler decide what to do next? In this case, it just looks at the value
of what’s in the brackets; you’ve set a to 0, so the value inside the
brackets is o.
In C, a value of o is equivalent to false, and a non-zero value is
equivalent to true. So by replacing the double equals with a single
equals, you’ve changed the value of a, and then you look to see if the
18
IMat/Pi [ Chapter Three ]
[ LEARN TO CODE WITH C ]
teste - /home/pi - Geany
File Edit Search View Document Project Build Tools Help
G? v IB v |S>x ^ v 51
□ □□
4 Cl
(P Functions
1
2
#include <stdio.h>
(p main [3]
3
void main (void)
4 E
{
S
int a = 0;
6
|| 7
if (a == 0)
8 E
t
9
printf (“a is equal to 0\n“);
10
>
11
else
12 4
i
13
printf ("a is not equal to 0\n");
14
n
15
}
16
- -- —■•'—
Status
15:31:02: File /home/pi/math.c saved.
15:31:26: File /home/pi/math.c saved.
Compiler 17:19:13: File /home/pi/math. c saved.
-^- 17:19:26: File /home/pi/test, c saved.
line: 14/16 col:5 sel:0 INS TAB mode: Unix (LF) encoding: UTF-8 filetype: C scope:main
value you’ve set a to is equivalent to true or false; neither of which
were what you wanted to do! If a C program is behaving strangely,
check very carefully that all your tests are actually tests and not
assignments: this is a very easy mistake to make.
So == is the test to see if a value is equal to another one.
There are other useful symbols that can be used in a test. The symbol
! =, for example, means ‘is not equal to’. The mathematical operators >
and < are used to test for ‘is greater than’ and ‘is less than’ respectively,
and they can also be combined with an equals sign to give >= and <=, the
tests for ‘is greater than or equal to’ and ‘is less than or equal to’.
You can combine tests with logical operators. The symbol && is a
Boolean AND (i.e. test whether both sides are true), and | | is Boolean
OR (i.e. test if either side is true). So, to execute code only if both a and
b are 0, you would use if (a == 0 && b == 0). To check if either
a or b is 0, you use if (a == 0 || b == 0)
Similarly, you can use the operator ! as a Boolean NOT to invert the
result of a test, so if (!(a == 0) ) is the same as if (a != 0) .
Above Make sure
that you use a
double equals sign
in the brackets
after the if, not
a single one!
I Conditions & Comparisons )
19
[ INFINITE
i nnPR l
III
Looping
The if statement is useful for making a single decision, but what if you
want to do something repeatedly until a test is true or false? We use a
while loop for this, and here’s an example:
#include <stdio.h>
void main (void)
{
int a = 0;
while (a < 5)
{
printf ("a is equal to %d\n", a);
a++;
}
printf ("a is equal to %d and I've finished\n", a);
}
This is very similar to an if statement, but the code in the curly
brackets is executed repeatedly for as long as the test in the round
brackets is true, not just once.
So in our example code, a is initialised to o. We enter the while
loop, and test to see if a is less than 5, which it is, so the code inside
the curly brackets is executed. The value of a is printed out, then we
have one of C’s useful shortcuts to save too much typing...
a++ is the same as a=a+l; the double plus means ‘add one to this
variable’. Similarly, a - - means ‘subtract one from this variable’; these
are very commonly used to count the times around a loop. The notation
a+=l can also be used to add a value to a variable; this also works for
other arithmetic operators, so a*=3 multiplies a by 3, and so on.
In the while loop, each time the code in the curly brackets has been
executed, the test in the round brackets is repeated; if it’s still true,
the loop code is repeated again. As soon as the test is false, execution
continues with the line after the closing curly bracket.
Sometimes, we might want a loop which always runs at least once
before a test is made. We do this with a small modification to the syntax
to create a do-while loop:
20
IMat/Pi [ Chapter Three ]
[ LEARN TO CODE WITH C ]
#include <stdio.h>
void main (void)
{
int a = 0;
do
{
printf ("a is equal to %d\n"j a);
a++;
} while (a < 5)j
printf ("a is equal to %d and I've finished\n", a);
>
The keyword do now goes before the curly bracket, and the while
and test go after the closing curly bracket. When this runs, the code
in the loop always executes once before the test; you can test this by
running both the loop examples above with a initialised to 5 rather
than 0, and seeing how the behaviour differs.
[ MORE
ABOUT
SEMICOLONS ]
Unlike the
test in an if
statement or a
while loop, you
need to put a
semicolon after
the test in a do-
while loop. This
indicates the
end of the loop
code; in a while
loop, the loop
code doesn't
end until the
last statement
inside the
curly brackets.
In the next chapter, we’ll look at some more complex examples
of looping and flow control.
Left A loop
executes the same
code multiple
times until the loop
test is false
[ Conditions & Comparisons 1
21
ESSENTIALS
CHAPTER FOUR]
MORE ADVANCED
FLOW CONTROL
For loops and case statements - more advanced
ways of controlling the flow of a program
[ LEARN TO CODE WITH C ]
Menu J© i ■*© - pi@raspberry: ~ II ijg'ifor.c - /home/pi - Gea... j ijpcase c - /home/pi -G._
'^p for.c - /home/pi - Geany
□BE
File Edit Search View Document Project Build Tools Help
SB [(_
'll
Symbols |>
ip Functions
i (vdd)
prim f ("a is equal t
J Symbols
(p Functions
St tus
Coi npiler
14:20i58: This Is Geany 1.24.1.
14:20:58: New tile "untitled'’ opened.
14:29:25: File /home/pi/for.c saved.
'•Jp case.c - /horne/pi - Geany (new instance)
File Edit Search View Document Project Build Tools Help
BEE
case.c x |
1 #include <stdio.h>
printf ("a is equal to l\n")
—printf—(‘a is greater than 1
equal to 5 and I've finished
pi@raspberry:~ $ gcc -o myprog case
piflraspberry: - S ./myprog
A for loop aLlows you to initialise,
test, and increment the variables
associated with a loop from
within the loop definition itself
J 15:01:02: This is Geany 1.24.1.
15:01:02: New file "untitled" opened,
15:02:53: File /home/pi/case, c saved.
A switch statement allows
you to choose different
actions depending on multiple
different values of a variable
[line: 15/16 col: 1 sel:0 INS TAB mode: Unix (LF) encoding: UTF-8 filetype. C scope: n
J
#include <stdio.h>
O The if statement and while loop described in the previous
chapter are fairly simple control structures. In this chapter,
we’re going to look at a few more complex structures that can
help to make your code shorter and reduce the amount of typing you
need to do...
While the while loop we saw in the previous article is very useful,
the for loop tends to be favoured by many programmers, as it puts all
the logic controlling the loop in one place. Here’s an example:
void main (void)
{
int a;
for (a = 0; a < 5; a++)
{
printf ("a is equal to %d\n", a);
}
printf ("a is equal to %d and I've finished\n", a);
I
[ More Advanced Flow Control ]
23
ESSENTIALS
[ MULTIPLE
INITIALISATIONS ]
You can
initialise
multiple
variables in a
for loop - just
separate them
by commas.
So if you want
to set two
variables at
the start of
the loop, you
can use for
(a = 0, b =
lj<test>;
<increment>)
r
This isn’t all that different from a while loop, but all of the control
for the loop lives in the round brackets after the for keyword. This
contains three statements, separated by semicolons: in order, these
are the initial condition, the test, and the increment.
a = 0 is the initial condition; the variable a is initialised to o at the
start of the loop.
a < 5 is the test, just like in a while loop. This is checked on each
iteration of the loop, and the loop code is only executed if the test
evaluates to true; as soon as the test is false, execution continues after
the curly bracket at the end of the loop code.
a++ is the increment; this is code which is executed at the end of
each iteration of the loop, before the test is evaluated again. In this
case, it adds l to a.
So when this for loop runs, what happens? First, a is set to o. The
test is then checked: is a (which is o) less than 5? Yes it is, so the code
inside the curly brackets is executed, and the value
pi@raspberrypi:
■
File Edit Tabs Help
pi@raspberrypi:~ $ ,/myprog
a is equal to 0
a is equal to 1
a is equal to 2
a is equal to 3
a is equal to 4
a is equal to 5 and I've finished
piQraspberrypi:- $ |
□ □□
The output when the for loop runs is
identical to that of the while loop in
the previous chapter - they both do
exactly the same thing
If [ Chapter Four ]
[ LEARN TO CODE WITH C ]
of a is printed. Finally, the increment is applied, meaning 1 ^
is added to a. ^
The test is then repeated. If true, the loop code is executed again,
and the increment is again applied; this repeats over and over until
the test is false, at which point execution continues after the closing
curly bracket.
In terms of what they do, for loops and while loops are pretty
much identical; both wrap up a section of code you want to run more
than once in some logic that controls how many times it runs. You can
use whichever makes the most sense, or whichever looks tidiest to you!
Switch statements
One thing that you quite often want to do is to test a variable against
several values, and do different things based on each of them. You can
do this with a set of nested if statements:
ftinclude <stdio.h>
void main (void)
{
int a = 0;
if (a == 0)
{
printf ("a is equal to 0\n");
}
else if (a == 1)
{
printf ("a is equal to l\n");
}
else
{
printf ("a is greater than l\n")j
}
>
That does start to get pretty long-winded, though, so C provides
a neater way of doing this, called a switch statement.
I Mors Advanced Flow Control ] 'Ma^Pi 25
#include <stdio.h>
void main (void)
{
int a = 0;
switch (a)
{
case 0 :
case 1 :
default :
>
printf ("a is equal to 0\n")j
break;
printf ("a is equal to l\n");
break;
printf ("a is greater than l\n");
>
This does exactly the same as the example above with multiple
if statements, but is a lot shorter. So how does it work?
The opening line consists of the keyword switch, with the name
of a variable in round brackets. This is the variable which will be
tested against the various cases.
The body of the switch statement is a number of case statements.
The variable a is compared against each case in turn; if it matches the
value immediately after the word case, then the lines of code after
the colon are executed.
The final case is just called default - every switch statement
should include a default case as the final one in the list, and this is
the code which is executed if none of the other cases match.
Notice that the last line in each case section is the word break -
this is very important. The keyword break tells the compiler that
you want to “break out” of the switch statement at this point; that
is, to stop executing code inside the switch and to resume execution
after the closing curly bracket. If you forget to include the break
statements, every case after the one you wanted will execute as
well as the one you wanted. Try it by compiling the code above
and running it - you’ll see the following in the terminal:
[ LEARN TO CODE WITH C ]
a is equal to 0
Now remove the two break statements, so the switch looks like:
switch (a)
{
case 0 : printf ("a is equal to 0\n");
case 1 : printf ("a is equal to l\n");
default : printf ("a is greater than l\n")j
}
and run it again - you’ll now see:
a is equal to 0
a is equal to 1
a is greater than 1
Not what you expected! This is another common bug in C code -
forgetting the break statements in your cases can result in very unexpected
Left Don't
forget the break
statements at the
end of each case
in your switch
statements!
1
[YOUR
FAVOURITE
LOOP... ]
All three types
of Loop in C -
while, do-while
and for - can
be used in
pretty much
any situation
where a loop
is needed:
choose
whichever
you like. Some
people prefer
to use one
type of loop
for everything:
others pick
and choose
whichever
looks tidiest
for each
circumstance.
There are no
wrong choices!
[ More Advanced Flow Control ] Mck/Pi
27
behaviour. But this can also be useful; programmers will sometimes
structure a switch statement with code that they want to execute in
multiple different cases, and deliberately leave out the break statements.
Leaving a loop early
The break statement has one other use; it can be used inside while
and for loops to break out of them. Look at this example:
#include <stdio.h>
void main (void)
{
int a = 0;
[CONTINUE]
The keyword
continue can
be used in a
loop instead
of break, but
instead of
breaking out
of the loop,
continue
skips all
the rest of
the code in
the current
iteration, and
returns to the
test case at
the start of the
loop. Among
other things,
this can be
useful to speed
up your code.
while (1)
{
printf ("a is equal to %d\n", a);
a++;
if (a == 5)
{
break;
>
}
printf ("a is equal to %d and I've finished", a);
}
So we have a while loop in which the test is just the value 1; this is
a non-zero value, and so is always true. If you enclose code inside curly
brackets after a while (1) statement, the loop will never end; it will
keep running forever. But in this case we have provided an alternative
way to end the loop; we test the value of a inside the loop itself in an
if statement, and if a is equal to 5, we call break. This causes the loop
to end and execution to continue with the statement after the loop. A
break statement like this can be useful to leave a loop early in the event
of an error, for example.
That’s about all you need to know about control flow in C; in the next
chapter, we’ll look at pointers, which are one of C’s most useful and
powerful features.
28 I MatfPi [ Chapter Four ]
CHAPTER FIVE
POINTERS
ESSENTIALS
Q : ■ * O •*»)*point.c - /home/pi
I pi@raspberry: -
[ * AND &]
When I was
first learning
about pointers,
I found it
helpful to say
out loud what
a line of code
was doing - an
* is "what is
pointed to by",
and an & is "the
address of’.
Once you've
got those
two ideas
fixed in your
head, you've
pretty much
understood
pointers!
O he term pointer has struck fear into the heart of many a
beginner C programmer, but once you’ve got your head around
them, they are a very useful feature of the language. They
aren’t actually that complicated in reality, but it’s easy to get confused
when using them, so let’s try to avoid that...
Remember when we looked at the declaration of variables? Declaring
a variable - telling the compiler what type it is and what it’s called -
before you can use it is necessary in C, because the declaration enables
the compiler to allocate a block of memory to store the variable. So for
every variable you declare, there’s a block of memory which is set aside
by the compiler for that variable, and the compiler remembers which
particular block of memory is used for each variable.
What is a pointer?
A pointer is just the address of a block of memory with a variable in it;
that’s all there is to it. So if you declare a variable and a pointer to that
variable, you can access the value in that block of memory in two ways;
either with the variable name, or with the pointer.
[ Chapter Five ]
[ LEARN TO CODE WITH C ]
Let’s look at a simple example:
#include <stdio.h>
void main (void)
{
int a;
int *ptr_to_aj
ptr_to_a = &a;
a = 5;
printf ("The value of a is %d\n", a);
*ptr_to_a = 6;
printf ("The value of a is %d\n", a);
printf ("The value of ptr_to_a is %d\n"j ptr_to_a)
printf ("It stores the value %d\n", *ptr_to_a);
printf ("The address of a is %d\n", &a)j
>
Taking it line by line, the first line is one we’re already familiar with:
we declare an integer variable called a. But what’s this?
int *ptr_to_a;
This looks like it’s declaring another integer variable, doesn’t it?
But look more carefully; the asterisk (*) at the start of the variable name
indicates that this is not declaring an integer variable, but a pointer to an
integer variable.
So we now have an integer variable called a, and we have a pointer to
an integer variable, called ptr_to_a. But neither of these actually have
a value in them yet: they are both uninitialised. It’s all very well calling
the pointer ptr_to_a, but it has no idea what (or where) a is, so we’ll
fix that with the next line.
mi
ptr_to_a = &a;
[ Pointers)
31
ESSENTIALS
[ MEMORY ]
Pointers are
one of the
ways C allows
(or in some
cases forces)
you to think
about what
the actual
hardware
of your
computer is
doing - a good
understanding
of pointers
gives you
a good
understanding
of how the
compiler
handles
memory.
This is the important bit! In C, the symbol & before a variable name
means address of the variable, so &a means “the address in memory
of the variable a”. And as we said above, a pointer is the address of a
variable. So this line initialises ptr_to_a to be the address of a; ptr_
to_a is now a valid pointer to the variable a, so we can now use it.
The next two lines are familiar; we set a to be 5, and just to check
that worked, we print its value. So let’s try doing the same thing,
but with the pointer.
*ptr_to_a = 6;
That asterisk again, but used in a slightly different way from
before. When declaring a variable, putting an asterisk before its name
indicates that the variable is a pointer. But once the pointer exists,
putting an asterisk in front of its name means the variable pointed
to by this pointer; this is known as dereferencing the pointer. So this
line tells the compiler to set the variable pointed to by the pointer
ptr_to_a to 6. We know that the variable pointed to by ptr_to_a is
a; we set that up a couple of lines back, and so this line is just another
way of setting a to 6; indeed, if we print the value of a, we find it has
changed to 6.
The next lines will hopefully help you get the relationship between
pointers, variables, and addresses clear in your mind.
printf ("The value of ptr_to_a is %d\n"j ptr_to_a);
In this line, we’re printing the value of ptr_to_a; not the
value it’s pointing at, but the value of the pointer itself. This
prints a very large number, as it’s the address in memory where
a can be found.
printf ("It stores the value %d\n", *ptr_to_a);
In this line, we’re printing the value pointed to by ptr_to_a; note
the asterisk before the name. This prints the value of a.
printf ("The address of a is %d\n", &a)j
32
MatfPi [ Chapter Five 1
[ LEARN TO CODE WITH C ]
Finally, in this line we’re printing the address of a itself;
note the & sign before the name. Again, this prints a very large
number, the same as the value of ptr_to_a we printed above.
The crucial thing to remember when working with pointers is
this: you can’t just declare a pointer, as you need to also declare
and associate the variable you want it to point to. When a pointer is
created, it points at a random location in memory; if you try and write
something to it, you can cause all sorts of errors up to and including
crashing the computer completely! Always make sure your pointers
are pointing at something before doing anything with them.
it, this makes sense; a pointer is just an address in memory, so we
don’t necessarily need to know what is at that memory. To use a void
pointer, you need to cast it; you tell the compiler what sort of pointer
to treat it as, and what sort of variable is in memory at that location.
Here’s an example:
[ Pointers) 'MacfPi
33
#include <stdio.h>
void main (void)
{
int intval = 255958283;
void *vptr = &intval;
printf ("The value at vptr as an int is %d\n"j *((int *) vptr));
printf ("The value at vptr as a char is %d\n"j *((char *) vptr));
>
You can use
++ and - - on
pointers, but
you need to
be careful.
(*a)++
increments
the value
pointed to by
a, but *(a++)
increments
the pointer
itself rather
than the value
it points at -
this will move
a to point at
the memory
address
immediately
after a.
We initialise the void pointer vptr to point to an integer variable
called intval.
In the first printf statement, we insert (int *) in front of vptr
before we dereference it using *. This casts vptr to an integer pointer,
and so the value of intval is printed as an integer.
In the second printf statement, we insert (char *) in front of vptr
before we dereference it. This casts vptr to a char pointer, and so what’s
printed is the value of the char which makes up the first byte of intval.
What do you use pointers for?
That’s really all there is to pointers, other than to ask why bother?
We can already access a variable with its name, so why do we need
to have some complicated alternative way of getting to the contents
of a variable?
There are several ways in which pointers are useful, which we will
explore in more detail in later chapters. But a few of the important
ones are:
> function calls - in the next chapter we will look at how to split up C
code into functions; pointers are very useful for allowing a function
to return multiple values.
> string handling - in C, a string is a continuous block of memory with
a letter stored in each byte; pointers make it possible to perform
efficient operations on strings.
> arrays - C allows array variables, lists of values of the same type,
which like strings are stored in a continuous block of memory;
pointers make accessing arrays easier and more efficient.
ESSENTIALS
CHAPTER SIX
FUNCTIONS
Functions - how to split your code
up into easy bite-sized chunks...
© Lffll
•gpfunc.c - /home/pi - Ge.. M
pi@raspberry:
[ ARGUMENTS]
A function
can have any
number of
arguments,
from zero up
to hundreds.
If you don't
need any
arguments,
you list the
arguments
as (void) in
the function
definition (just
like in the main
function);
when you call
the function,
just put a
pair of empty
round brackets
() after the
function name.
O p until now, all the examples we’ve looked at have had one
single function, main, with all the code in it. This is perfectly
valid for small, simple programs, but it’s not really practical
once you get more than a few tens of lines, and it’s a waste of space if
you need to do the same thing more than once. Splitting code up into
separate functions makes it more readable and enables easy reuse.
We’ve already seen functions used; the main function is a standard
C function, albeit with a special name. We’ve also seen the printf
function called by our examples. So how do we create and use a
function of our own? Here’s an example:
#include <stdio.h>
int sum (int a, int b)
{
int res;
res = a + b;
return res;
>
k,
36 I MatfPi [ Chapter Six ]
void main (void)
{
int y = 2;
int z = sum (5^ y);
printf ("The sum of 5 and %d is %d\n" J y, z);
>
This includes both the main function and a second function called sum.
In both cases, the structure of the function is the same: a line defining
the value returned by the function, the function name, and the
function arguments, followed by a block of code enclosed within curly
brackets, which is what the function actually does.
What's in a function?
Let’s look at the sum function:
int sum (int a , int b)
The definition of a function has three parts. The first part is the type
of the value returned by the function: in this case, an int. The second
part is the name of the function: in this case, sum. Finally, within round
brackets are the arguments to the function, separated by commas, and
each is given with its type: in this case, two integer arguments, a and b.
The rest of the function is between the curly brackets.
int res;
This declares a local variable for the function, an integer called res.
This is a variable which can only be used locally, within the function
itself. Variables declared within a function definition can only be used
within that function; if you try and read or write res within the main
function, you’ll get an error. (You could declare another int called
res within the main function, but this would be a different variable
called res from the one within the sum function, and would get very
confusing, so it’s not recommended!)
[VARIABLE
SCOPE]
If you declare a
variable within
a function, it’s
only usable
within that
function, not
within any
functions
which call
the function,
or within
functions
called by the
function. This
is known as
the scope of a
variable: the
parts of the
code in which
it’s valid.
[ Functions ] MacfPi
37
res = a + b;
This should be obvious! Note that a and b are the two defined arguments
of the function. When a function is called, a local copy of the arguments
is made and used within the function. If you change the values of a or
b within the function (which is a perfectly valid thing to do), that only
affects the value of a and b within this function; it doesn’t change the
values that the arguments had in the function from which it was called.
return res;
Finally, we need to return the result. The function was defined to
return an integer, so it must call the return statement with an integer
value to be returned to the calling function.
A function doesn’t have to return a value; if the return type
is set to void, it returns nothing. There’s no need for a return statement
in a function with a void return type; the function will return when it
reaches the last line; however, if you want to return early (in the event
of an error, for example), you just call return with no value after it.
Calling a function
The sum function returns an integer, so we set an integer variable
equal to it. The arguments we supply to the function are inside round
brackets, and in the same order as in the function definition; so in this
case, a is 5, and b is the value of y.
Can you return more than one result from a function? You can only
return one value, but you can also use pointers to pass multiple items
of data back to the calling function. Consider this example:
ttinclude <stdio.h>
int sum_and_diff (int a, int b, int *res)
{
int sum;
sum = a + b;
*res = a - b;
return sum;
}
void main (void)
{
int b = 2;
int diff;
printf ("The sum of 5 and %d is %d\n", bj
sum_and_diff (5, b , &diff));
printf ("The difference of 5 and %d is %d\n", b, diff);
>
We’ve modified the sum function to calculate both the sum
and the difference of the arguments. The sum is returned as before,
but we’re also passing the difference back using a pointer. Remember
that the arguments to a function are local variables; even if you
change one in the function, it has no effect on the value passed by the
calling function. This is why pointers are useful; by passing a pointer,
the function doesn’t change the value of the pointer itself, but it can
change the value in the variable to which it’s pointing.
So we call the function with the same two arguments as before, but
[ Functions ]
39
[MODIFYING
ARGUMENTS]
we add a third one, a pointer to the variable where we want to write
the difference calculated by the function. In the function, we have
Arguments are
local variables
within a
function. If
you want a
function to
modify the
arguments
you give it,
make each
argument you
want to modify
a pointer to
a variable;
you can then
read the value
pointed to
this line:
♦res = a - b;
The difference is written to the variable to which res is a pointer.
In the main function, we call the sum_and_dif f function like this:
sum_and_diff (5, b, &diff)
We provide the address of the integer diff as the pointer argument
to the sum_and_diff function; when the difference is calculated, it’s
written into the variable diff in the main function.
within the
function, and
write the
changed value
back to the
same pointer.
Right By using
a pointer as one
argument, the
sum_and_diff
function can
return both
the sum and
difference of the
arguments
Order matters
One thing to bear in mind when defining functions is that the compiler
reads files from top to bottom, and you need to tell it about a function
before you can use it. In the examples above, this is automatic, as the
definition of the sum and sum_and_diff functions is before the first
call to them in main.
40
MatfPi [ Chapter Six ]
[ LEARN TO CODE WITH C ]
:? func.c - /home/pi - Gesny
File Edit Search View Document Project Build Tools Help
X
SB
4
<J Symbols >
- Functions
<P main [ 11 ]
(p sum_and_diff [
func.c x |_
1 #include <stdio.h>
2
3 int sum and diff (int a, int b, int *res)
t?{
int sum;
sum = a + b;
*res = a - b;
return sum;
>
void main (void)
int b = 2;|
int diff;
printf ("The sum of 5 and %d is %d\n", b, sum_and_diff (5, b, Sdiff)
printf ("The difference of 5 and %d is %d\n", b, diff);
Status
Compiler
Messages
10:49:09: File /home/pi/sum.c saved.
10:49:16: File /home/pi/point, c saved.
10:54:56; File /home/pi/point.c saved.
10:55:04: File /home/pi/func.c saved.
10:55:16: File /home/pi/func.c saved.
line: 13/19 col: 14 sel.O INS TAB mode: Unix (LF) encoding: UTF-8 filetype: C scope: main
But in larger files, when multiple functions call multiple other
functions, this gets complicated; it’s not always easy to make sure the
function definitions are all in the right order. To avoid this, C allows
you to declare functions before they are used.
A function declaration is just the definition of the function, minus
the function code within the curly brackets. So for the sum_and_diff
function, the declaration would be:
int sum_and_diff (int a, int b , int *res);
Note the semicolon at the end! Function declarations are included at
the top of the file; when the compiler finds a function declaration, it
knows that at some point a function with this name, arguments, and
return type will be defined, so it then knows how to handle a call to it,
even if it hasn’t yet seen the definition itself.
I ■
[ Functions 1 IMacfPi
ARRAYS &
STRINGS
How to handle arrays (lists of values)
and strings (lists of letters) in C
[ LEARN TO CODE WITH C ]
Menu |
3 b ■ * ©!
I pi@raspberry>
I' tjpairay.c - /home/pi - G...
<1 Symbols |>
(p Functions
■•JP array.c - /home/pi - Geany
EBB
File Edit Search View Document Project Build Tools Help
C?
& *
4 a
array.c x
#include <stdio.h>
void main (void)
int a[ 10 ];
int count;
(count = 0; count < 10; count++)
a[count] = count * 10 + count;
printf ("The first and second elements of a
printf ("Or, as pointers, %d and Sd\n", *a.
are %d and %d\n" ,
*(a+l»;
IS
I); |FHe
to], a[i]); File Edit Tabs Help
aspDerry: > gcc -o myprog array.c
pigraspberry$ ./myprog
The first and second elements of a are 0 and 11
as pointers, 0 and 11
pigraspberry:- $ |
An array declaration, showing
the type, the name, and the
number of elements
An array can be accessed
by its indices inside square
brackets, or as a pointer
int a[10];
O he variables we have looked at so far are all single numeric
values. In this chapter, we’re going to look at how C handles
lists of values, and that leads into using lists of letters to store
and manipulate text strings.
An array is a single variable which stores multiple different values of the
same type; the individual values are accessed by indexing the array. An array
can have one or more dimensions-, a one-dimensional array is a single list of
values, while a two-dimensional array is a list of lists of values, and so on.
An array is declared in C by putting the size of each dimension in
square brackets after the variable name. So
is a list of to integers, while
int b[ 5 ][6];
is a list of 5 lists, each of which contains 6 integers.
When accessing the elements inside an array, the array index - the
number inside the bracket - starts at o. So the to integers contained within
array a above are referred to as a [0], a [1], a [2] and so on up to a [9], The
compiler will quite happily allow you to read or write a [ 10 ], a [ 11 ] or indeed
a [ any number you like ], but these are all outside the memory which was
allocated when the array was declared, so writing to them is a really bad idea!
I
[ Arrays & Strings ]
43
ESSENTIALS
[ KEEP INSIDE
YOUR ARRAY]
One of the
nastiest
sources of
crashes and
bugs in C is
creating an
array and then
writing past
the end of it.
The compiler
won't stop
you writing
to memory
off the end of
an array, and
doing so can
have serious
consequences.
Always make
sure your
array indices
fit inside
your array.
Arrays and pointers
This brings us on to the relationship between pointers and arrays.
The name of an array is effectively a pointer to the first element of the
array. Remember that a pointer is the address of a variable in memory?
Well, an array is a contiguous block of memory which contains all the
elements of the array in order, so you can use a pointer to access it. (In
fact, even if you use values in square brackets to access it, the compiler
treats those as a pointer anyway.) Here’s an example:
#include <stdio.h>
void main (void)
{
int a[10];
int count;
for (count = 0; count < 10; count++)
{
a[count] = count * 10 + count;
>
printf ("The first and second elements of a are %d and %d\n",
a[0L a[l])i
printf ("Or* as pointers, %d and %d\n", *a, *(a+l));
>
This fills the to values of a with the numbers 0, u, 22, 33, and
so on, and then reads a[ 0 ] and a [ 1]. It then reads the same values
using a as a pointer, and you can see if you run the code that they
are identical.
With a two-dimensional array or greater, you need to consider
how the compiler arranges the dimensions in memory; it does so by
grouping the elements at the rightmost index of the array together.
With the array b[ 5] [6] above, b itself points at b[0] [0]. b+1 points
at b[0] [1]; b+5 points at b[0] [5]; and b+6 points at b[ 1] [0].
You can initialise an array at the same time as you declare it by
putting the values in curly brackets, so:
44
IMat/Pi [ Chapter Seven ]
[ LEARN TO CODE WITH C ]
int a[10] = { 0, 11., 22, 33, 44, 55, 66, 77, 88, 99 };
>;1
But note that this only works when the array is first declared; once it
exists, you can’t use this shortcut and will need to iterate through the
array indices, setting each value in turn.
a[0]
a[1]
a [2]
a[3]
a[4]
a[5]
a [6]
aft]
a[8]
a[9]
b
t
b[0][0]
b[0][1]
b[0][2]
b[0][3J
b[0][4]
b[0][5]
b[1][0]
b[1][1]
b[1][2]
b[1][3]
Above Array elements are stored sequentially in memory, with the array name a
pointer to the first element. Multi-dimensional array elements are stored with the
elements with neighbouring values, in the rightmost index next to each other
Strings
In C, a string is just another array; it’s an array of single characters. A character
is a specific type in C, called chap; this holds a single byte, which is enough to
Or, to initialise it at the same time:
char mystring[10] = "thestring";
[NAMES ARE
POINTERS ]
Remember
that the name
of an array or a
string is just a
pointer to the
first element
of the array
or string in
question, and
can be used
in the same
way as any
other pointer;
it can be
incremented
and
decremented,
or
dereferenced
to find the
value to which
it points.
One important thing to remember is that a string in C must always
end with a byte set to zero, and that the memory required to hold this
final zero (called the string terminator ) must be allocated when you
declare the string. So mystring, which is declared as an array of to
chars, can only actually hold text of 9 or fewer letters.
Below Strings
are stored as an
array of single
characters, with
the element after
the last character
set to zero
mysfring
‘t’
'h'
‘e’
‘s’
‘t’
‘r*
‘i’
‘n’
‘g’
0
[ Arrays & Strings ] 'Ma^Pi
45
ESSENTIALS
You can use the index in square brackets to access individual
characters in a string, or you can use a pointer. Here’s an example
of using pointers to join two strings together:
#include <stdio.h>
void main (void)
{
char strl[10] = "first";
char str2[10] = "second";
char str3[20];
char *src, *dst;
[ TERMINATING
STRINGS]
Always
remember that
the memory
you allocate
for a string
needs to be
long enough
to hold all the
characters,
plus one extra
to store the
terminating
zero. If you're
manipulating
strings yourself
with pointers,
make sure you
remember to
write the zero
at the end
of any string
you create.
src = strl;
dst = str3;
while (*src != 0)
{
*dst = *src;
src++;
dst++;
}
src = str2;
while (*src != 0)
{
*dst = *src;
src++;
dst++;
}
*dst = 0;
printf ("%s + %s = %s\n"., strl, str2, str3);
>
First, we create two strings - strl is “first” and str2 is “second”
and we allocate an empty string, str3, to put the result in.
We then create a pair of char pointers, and point src at the
start of strl (the “f” of “first”) and dst at the start of the empty
[ Chapter Seven ]
[ LEARN TO CODE WITH C ]
str3. We then loop, copying what’s at src to dst, and then
moving both pointers forward by one, until we find the zero that
terminates strl.
We then point src at str2, and do the same thing again, until we
find the zero at the end of str2. Finally, we write a zero to the end
of str3 to terminate it. Note the new format specifier used to print
strings; %s is used to print a string, and will display every character
from the pointer supplied as an argument, up to the first terminating
zero it finds. (If you want to print a single character, you can use the
format specifier %c.)
Writing to strings
Because the name of a string variable is only a pointer to the first
character of the string, you can’t just use an equals sign to set the
value of a complete string. You can initialise a string variable at the
time you declare it, as above, but what if you want to set or change
it later?
There are a few ways to do this, but the most useful is the sprintf
function; this is a version of the printf function we have already seen,
which writes arbitrary text to string variables. The only difference
is that the first argument it takes is the name of a string variable,
and it writes to that instead of to the terminal:
[ KEEP INSIDE
YOUR STRING ]
As with arrays,
it's even easier
to write off the
end of a string
and to corrupt
memory
elsewhere as
a result. If in
doubt, declare
your string
variables with
more space
than you think
you'll need,
and make
sure that what
you're writing
into them
actually fits.
#include <stdio.h>
void main (void)
{
int val = 12;
char string[50];
sprintf (string, "The value of val is %d\n", val);
printf ("%s"j string);
>
mi
The sprintf function will automatically add the terminating zero
at the end of any string you create with it.
In the next chapter, we’ll look at some of the functions provided
in C’s string handling library to make working with strings easier.
[ Arrays & Strings ] JWat/Pi
47
I MadPi
ESSENTIALS <j
CHAPTER EIGHT
THE STRING
LIBRARY
Using the C string library to simplify
common operations on strings
[ LEARN TO CODE WITH C ]
© Hi
I pi(S)raspberry: -
'•jpstring.c - /home/pi -...
15:43:16: File /home/pi/string.c opened(2).
[ KEEP INSIDE
YOUR STRING ]
The string
library
functions won't
generally
prevent you
from writing
off the end of
a string: just
as when using
pointers, when
using library
functions you
still need to
make sure
your string
variables are
large enough
for the values
you’re writing
into them.
O n the previous chapter, we saw how to access strings using
pointers. This works perfectly well, and gives you a good
understanding of how pointers work, but it’s a bit long-
winded. Fortunately, C provides a library of useful string functions,
which save a lot of typing!
In the last chapter, we saw how to join two strings together using
pointers. We’re going to do the same thing using the string handling
library. Here’s the code rewritten using string library functions:
#include <stdio.h>
#include <string.h>
void main (void)
{
char strl[10] = "first";
char str2[10] = "second";
char str3[20];
strcpy (str3, strl);
strcat (str3, str2);
printf ("%s + %s = %s\n"., strl, str2, str3);
>
That’s a lot shorter! Note the #include <string.h> at the start, which
tells the compiler we want to use functions from the string library.
This shows us two string functions, strcpy ('string copy’) copies
the string at the second argument to the start of string at the first
argument, strcat (‘string concatenate’) does the same thing, but
instead of copying to the start of the first argument, it finds the
terminating zero of the first argument and starts copying to its
location, thus joining the two strings together.
Comparing strings
Another common requirement is to be able to compare two strings
to see if they are the same. As we’ve already seen, we can compare
numeric values with the == operator, but this doesn’t work with
strings. Remember that the name of a string is really just a pointer
to a location in memory containing the string, so using == to compare
two strings will only tell you if they’re at the same place in memory,
not if two strings at different locations are the same.
You can use == to compare two char variables, and a string is an
array of chars, so it’s possible to write a simple piece of code that
compares each character in a string in turn:
#include <stdio.h>
void main (void)
{
char strl[10] = "first";
char str2[10] = "fire";
char *ptrl = strl, *ptr2 =
while (*ptrl != 0 && *ptr2
{
if (*ptrl != *ptr2)
{
break;
>
ptrl++;
ptr2++;
>
str2;
! = 0 )
[ LEARN TO CODE WITH C ]
if (*ptrl == 0 && *ptr2 == 0)
{
printf ("The two strings are identical. \n");
}
else
{
printf ("The two strings are different.\n") ;
}
>
But that’s a bit tedious to write out every time, so the string library
has done this for you with the function strcmp (for 'string compare’).
Here’s how you use it:
#include <stdio.h>
#include <string.h>
void main (void)
{
char strl[10] = "first";
char str2[10] = "fire";
if (strcmp (strl, str2) == 0)
{
printf ("The two strings are identical. \n");
}
else
{
printf ("The two strings are different. \n");
}
>
[ DON'T
OVERWRITE ]
It looks like it
ought to be
possible to
use strcpy
and strcat
to copy part of
a string over
itself - strcpy
(a + 1, a),
for example.
Don’t try it! The
source and
destination
buffers for
strcpy and
strcat must
be completely
separate areas
of memory;
if not, their
behaviour is
unpredictable.
strcmp takes two strings as arguments, and returns a 0 if they’re the
same; it returns a non-zero value if not.
What about if you just want to compare the first few characters of a
string, not the whole string? There’s a library function for that, too:
strncmp (for 'string numbered compare’).
[ the String Library ] IWacfPi
51
[ IGNORING
CASE]
I
There are
versions of
strcmp and
strncmp
which ignore
the case of the
letters in the
strings being
compared;
they're called
strcasecmp
and
strncasecmp,
respectively.
They take
the same
arguments
and return the
same values.
This works in exactly the same way as strcmp, but it takes a third
argument, an integer giving the number of characters to compare.
So strncmp ("first", "fire", 4) would return a non-zero value,
while strncmp ("first", "fire", 3) would return a o.
Reading values from a string
We saw in the previous chapter that we can use sprintf to write
variables into a string; what about being able to read variables back out
of a string? The function sscanf (‘string scan formatted’) does that for
you. Here’s how it works:
#include <stdio.h>
void main (void)
{
int val;
char string[10] = "250";
sscanf (string, "%d", &val);
printf ("The value in the string is %d\n", val);
>
sscanf uses exactly the same format specifiers as printf.
One important difference, though, is that the arguments to sscanf
must all be pointers to variables, rather than variables themselves. As
always, a function can never change the values of variables provided
as arguments, but it can write to their destinations if they are
pointers.
You can check whether sscanf was able to match the format
specifiers with the string provided by looking at the value it returns;
sscanf returns the number of values it successfully read. So for
example, if a format specifier of %d is provided but the string supplied
doesn’t start with a decimal number, sscanf will write nothing to the
supplied pointer and will return 0; if the string supplied does start with
a decimal number, sscanf will return l.
I
52
MatfPi [ Chapter Eight ]
[ LEARN TO CODE WITH C ]
The format string supplied to sscanf can contain multiple format
specifiers and even other text:
#include <stdio.h>
void main (void)
{
int val;
char result[10]j
char string[25] = "The first number is 1";
if (sscanf (string, "The %s number is %d", result, &val) == 2)
{
printf ("String : %s Value : %d\n", result, val);
}
else
{
printf ("I couldn't find two values in that string. \n");
}
>
Note that, slightly inconsistently, the %s format specifier denotes a
pointer to a string in both printf and sscanf, while the %d specifier
denotes a variable in printf but a pointer in sscanf.
One thing to note about sscanf is that it’s in the standard I/O
library, not the string handling library, so you don’t need #include
<string.h> to use it.
[SSCANF
STRINGS ]
The %s format
specifier
matches a
set of non¬
whitespace
characters in
sscanf: it will
extract the first
set of letters,
numbers or
punctuation
up to the first
space, tab or
newline it finds
in the string
being scanned.
A space in
the format
string matches
one or more
whitespace
characters,
not just a
single space.
53
[ The String Library ] ‘MacfPi
Above sscanf
reads numeric
values and words
out of a formatted
string, allowing
you to parse text
How long is a (piece of) string?
One final useful string handling function is strlen (for ‘string length’);
as the name suggests, this tells you how many characters there are in a
string, counting from the start to the terminating zero character.
from elsewhere.
Remember that
all the arguments
to sscanf must
be pointers
Sinclude <stdio.h>
ttinclude <string.h>
void main (void)
{
char strl[10] = "first";
printf ("The length of the string
strlen (strl));
'%s' is %d\n"j strlj
All the operations we’ve looked at here are possible by manually
manipulating pointers; the string library just makes them easier and
will make your code shorter. If you find yourself moving pointers
around strings in a program, always check the string library to make
sure you’re not reinventing the wheel!
54
IMat/Pi [ Chapter Eight ]
ESSENTIALS
CHAPTER NINE
USER INPUT
Reading and interpreting
input from the user
II
■ II
ESSENTIALS
ylenu
I pigiraspberrypi: -
ijf input.c - /home/pi - G..
itebasket
•« l inputc - /home/pi - Geany
File Edit Search View Document Project Build Tools Help
D?
la» x
<>i %
SB
stdin is the input stream
associated with the terminal
Symbols
cp Functions
>
(p main [3]
inputc x
#include <stdio.h>
void main (void)
[=H
char input[256], name[256];
int age;
21 L >
printf ("What is your name,
fgets (input, 256, stdin);
sscanf (input, "%s", name);
?\n" );
you?\n“, name);
printf ("Hello, %s. How old are
while (1)
{
fgets (input, 256, stdin),
if (sscanf (input, "%d" , &age) == 1) break;
printf ("That wasn't anything I recognise as an
}
pi(S)raspberrypi: -
File Edit Tabs Help
printf ("Well, %s, you look young for %d...\n", name
piQraspberrypi:-' $ gcc -o myprog inpi
piQraspberrypi :~ $ ./myprog
What is your name, user?
Simon
Hello, Simon. How old are you?
42
Well, Simon, you look young for 42..
piQraspberrypi : $ |
Status
Compiler
occanoc
14:36:03: File /home/pi/input.c opened(l).
14:37:20: File/home/pi/input.c saved.
14:37:49: File /home/pi/input.c saved.
|/ed.
fgets is the library function
to read text from an input
stream
mode: Unix (LF) encoding: UTF-8 filetype: (3
O n the previous chapter, we looked at how to print program
output to the terminal, but in order to interact with the user,
this needs to be a two-way process. This chapter looks at how
we can read and interpret input that the user enters in the terminal.
We’ve seen the printf function used a lot in previous chapters; it’s
the standard way of writing formatted text output from a program to
the console, the command line from which you run the program. But
what if you want to get input from the user? How do we read what the
user types into the console?
In the last chapters, we looked at the sscanf function which
reads values from a string. There’s an equivalent function called
scanf , which reads values directly from the console, as in the
following example:
56
HftcujPi [ Chapter Nine ]
[ LEARN TO CODE WITH C ]
#include <stdio.h>
void main (void)
{
char input[256]j
int age;
printf ("What is your name, user?\n");
scanf ("%s", input);
printf ("Hello, %s. How old are you?\n", input);
scanf ("%d". Sage);
printf ("Well, %s, you look young for %d_\n", input, age);
scanf works exactly like sscanf, but has one fewer argument,
as it reads from the console rather than from a string.
However, it’s not really the best way of getting console input; it
only really works if you have a user who types in exactly what you
expect. Unfortunately, users have a nasty tendency to type in things
you aren’t expecting, and scanf doesn’t cope well with this. For
example, in the code above, if the user types in 257 characters when
asked for their name, they will overflow the space allocated for the
input string, and bad things may happen...
ESSENTIALS
A better way
A better approach is to read each line the user enters into a
buffer string, and then use sscanf to read values from that
string. The C library function f gets is useful for this. Have a look
at this example:
#include <stdio.h>
void main (void)
{
char input[256], name[256];
int age;
[ STDIN
AND STDOUT]
We talk about
stdin in this
article, which is
the 'standard
input' stream:
what the
user types at
the console.
You may
sometimes see
references to
stdout, which
is the 'standard
output' stream
- as you might
expect, this is
output which
is printed to
the console,
usually
via printf.
printf ("What is your name, user?\n");
fgets (input, 256, stdin);
sscanf (input, "%s", name);
printf ("Hello, %s. How old are you?\n", name);
while (1)
{
fgets (input, 256, stdin);
if (sscanf (input, "%d", &age) == 1) break;
printf ("I don't recognise that as an age - try again!\n");
}
printf ("Well, %s, you look young for %d...\n", name, age);
>
fgets takes three arguments. The first is the buffer into which it
should store the input. The second is the maximum number of bytes
it will write into that buffer; this is useful to prevent the overflow
situation mentioned above. Finally, it takes an argument telling
it where to read from; in this case, this is set to stdin (short for
‘standard input’), which tells it to read from the console.
So each time we ask the user for input, we use fgets to read up to 256
characters of whatever they type (up to the point at which they press
the enter key), and we then use sscanf to interpret it. Additionally,
when asking for the user’s age, we use the value returned by sscanf
[ Chapter Nine ]
I
[ LEARN TO CODE WITH C ]
(described in the previous chapter) to check that the user has entered
what you expect, and loop until they give a valid answer. You can use
this method to interpret pretty much anything a user types, and to
safely handle all the cases where they type something unexpected!
Reading parameters
There’s another way to get input to your program, which is to supply
it as a parameter when you start the program from the command line.
At this point, I have to admit to not having been entirely honest for
the last 8 chapters... I’ve always shown the definition of the main
function as
void main (void)
This works, as you’ve seen, but it isn’t strictly correct. The strict
definition of main looks like this:
int main (int argc, char *argv[])
But let’s be honest: if I’d shown you that in chapter 1, you’d have run
a mile, wouldn’t you? So what does it all mean?
First off, we can see that main returns an integer; this is a
success or failure code which some operating systems can use for
processing in a shell script or the like. Traditionally, if a program succeeds,
main returns 0, and if it fails, it returns a non-zero error code. For
programs that run on their own, you really don’t need to worry about it!
What’s more useful are the other two arguments, argc is an integer,
and this is the number of parameters which were provided on the
[SCANF]
Just like
sscanf,
scanf returns
an integer
indicating how
many values
it successfully
read, which
you can use
to check for
errors. One
problem is
that scanf
only removes
matched
values from the
input buffer,
so if a scanf
fails to match
anything,
what the user
typed will be
read again on
the next call
to scanf. It
really is easier
to use fgets
and sscanf!
command line when the program was started. Strangely, the number
includes the program name itself, so this value is always l or more;
if parameters were provided, it will be 2 or more.
char *argv[ ]; now that’s confusing, right? This is actually a
composite of a few things we’ve already seen. There’s a * in there, so
it’s a pointer; the type is char, so there are characters in it, and there
are square brackets, so it’s an array...
This is actually an array of pointers to characters; each element of
the array is a string, and each string is one of the parameters provided
to the program.
mi
[ User Input ] MacfPi
59
ESSENTIALS
It’s probably easier to understand that in practice:
#include <stdio.h>
int main (int argc, char *argv[])
{
int param = 0;
while (param < argc)
{
printf ("Parameter %d is %s\n", param, argv[param]);
param++;
}
return 0;
>
Try running this as before, just by typing its name. Then try typing
other things after the name on the command line and see what the
program prints.
□□□
File Edit Search View Document Project Build Tools Help
60
IMat/Pi [ Chapter Nine ]
#include <stdio.h>
int main (int argc, char *argv[])
{
int argl, arg2;
if (argc == 4)
{
sscanf (argv[l]j "%d"j &argl);
sscanf (argv[3]j "%d"j &arg2)j
if (*argv[2] == ' + ') printf ("/SdXn",
if (*argv[2] == printf ("/SdXn",
if (*argv[2] == 'x') printf ("/SdXn",
if (*argv[2] == '/') printf ("/SdXn",
}
argl + arg2);
argl - arg2);
argl * arg2);
argl / arg2);
[ CHECKING
RETURN
VALUES]
In Linux,
the return
value from a
program isn’t
shown, but is
stored and can
be read from
the command
line. If you
type echo $?
immediately
after running
a program,
return 0 ;
>
Note that we use *argv[2] to get the first character of the second
parameter. This should only ever be a single character, but because
each of the arguments can be a string, argv[2] (without the asterisk)
is a pointer to a character, not the single character required for a
comparison using ==.
the value
the program
returned will
be shown.
Return values
are mainly
useful if
you're calling
programs
from scripts.
Make sure you separate the arguments from the operator with spaces
so they’re identified as separate parameters; <progname> 2 + 2
rather than <progname> 2+2
Left The calculator
reads the two
values and the
operator from the
a rgv array and
prints the result
llll
[ User Input ] MatfPi
61
ESSENTIALS <j
CHAPTER TEN
FILE INPUT
AND OUTPUT
Reading from and writing to files
[ LEARN TO CODE WITH C ]
Menu © : ■ * O M pi@raspberrypi - j ■_rtlmput.txt - /home/pi -...
O n the previous chapter, we looked at how to get input from
the user at the console. In this chapter, we’ll look at the
other common method of input and output in C: reading and
writing files.
Many programs need to be able to access files on the host computer’s
disk; even if it’s just for saving user preferences and the like, file access
is a fundamental requirement for a lot of programming tasks.
In C, files are accessed by use of file pointers. A file pointer contains
all the information required to access a file: both its name and location
on the file system, and the current position within the file at which
data will be read or written.
So the first thing we need to do is to get a file pointer. This is done
using the C function fopen, which takes two arguments. The first
argument is the path to the file, including its name and extension.
The second argument is called the file access mode-, this is a code which
indicates whether you intend to read from the file or write to it.
Let’s look at an example of reading a file. Use your text editor to
create a file called input. txt in the /home/pi directory on your Pi,
and type anything you like into it. Save it, and then create and run the
following program:
[ File input and output ]
63
ttinclude <stdio.h>
void main (void)
{
FILE *fp;
int value;
[ ALWAYS
CHECK YOUR
FILE POINTER]
Never assume
that f open
has worked -
always check
that the value
it returns is a
valid pointer
(i.e. not zero). If
you try to read
from a zero
pointer, you'll
get random
nonsense;
if you write
to a zero
pointer, you'll
probably crash
the computer!
fp = fopen ("/home/pi/input.txt"j "rb");
if (fp)
{
while (1)
{
value = fgetc (fp);
if (value == EOF) break;
else printf ("%c"., value);
}
fclose (fp);
}
>
First, we declare a file pointer variable called f p, which has the
type FILE *. We also declare an integer which we’ll use to hold the
characters read in from the file.
We then create the file pointer using the command fopen (for 'file
open’). We open the file at /home/pi/input .txt, and we set the mode
to "rb", which indicates ‘read binary’. This creates the file pointer and
initialises it to the beginning of the file.
We then check to see if the file pointer is non-zero; if the pointer is
returned as zero, the file wasn’t successfully opened. (For a read, this
usually indicates that the file doesn’t exist.)
If the file pointer does exist, we call the function fgetc (for ‘file get
character’) in a loop; each time this function is called, it reads a single
byte from the file, and then advances the file pointer to the next byte
in the file. When the file pointer reaches the end of the file, it returns
the special value EOF (for ‘end of file’). So we print the value returned
by fgetc each time until it returns EOF.
Once we have finished reading the file, we finish access to it by
64 I IMatfPi [ Chapter Ten ]
[ LEARN TO CODE WITH C ]
calling fclose (for ‘file close’), which frees the file pointer and
allows you to reuse it to access another file.
Note that while fgetc reads characters, it returns an integer;
this is because the code for EOF falls outside the valid range of a char
variable (o - 255). Unless at the end of a file, fgetc returns an integer
value which can always be treated as a char.
Writing a file
To write to a file, we use a file pointer in exactly the same way,
but we open it in a mode for writing.
#include <stdio.h>
void main (void)
{
FILE *fp;
int value;
fp = fopen ("/home/pi/output.txt"j "wb");
if (fp)
{
for (value = 48; value < 58; value++)
{
fputc (value, fp);
}
fclose (fp);
}
[ REMEMBER
TO FCLOSE ]
It's easy to
forget to call
fclose on
your file, but it’s
important to
do so. On some
systems, when
writing to the
file system, the
write doesn't
actually
complete
until fclose
is called; if
your program
doesn’t call
fclose, you
might find
that you write
to files and
nothing shows
up in them.
In this case, we open the file /home/pi/output.txt with the mode
"wb" , which indicates ‘write binary’. This opens the file for writing;
if this file already exists, the contents are deleted.
We then call the function fputc (for ‘file put character’) in a loop,
writing the bytes 48, 49...57 to the file. (These are the character codes
for the text characters for the 10 digits 0,1...9). As before, we then close
the file pointer. If you run this and then look in your home directory, you
should find the file output.txt, containing the string 0123456789.
[ Fils input and output ] 1Ua0 > i
65
Right Reading
or writing a file
requires a file
pointer to be
opened with
fopen, and the
resulting pointer
is then used in
all operations.
Remember to
close the pointer
afterwards
with fclose
[READ AND
WRITE THE
SAME FILE]
You can open
| X |
4\0l
#include <stdio.h>
void main (void)
value = fgetc (fp);
if (value == EOF) break;
else printf ("%c", value);
Formatted output
fputc is useful for writing bytes to a file, but it’s an inconvenient way
of writing text to a file. For this, we can use the f printf function ('file
print formatted’)•
a file for
simultaneous
reading and
writing with
the same
pointer. Set
the file access
mode to "rb+"
to read an
existing file and
to overwrite it;
set it to "wb+"
to create a new
file and be able
to read back
what you've
written to it; set
it to "ab+" to
open a file to
both append
at the end and
ffinclude <stdio.h>
void main (void)
{
FILE *fp;
fp = fopen ("/home/pi/output.txt", "wb");
if (fp)
{
fprintf (fp, "This is some text.\n");
fclose (fp);
}
>
read from it.
f printf works in exactly the same way as sprintf , but the first
argument is a file pointer rather than a string.
Moving around a file
Quite often, rather than overwriting a file, we just want to add to the
end of it. To do this, open it with fopen ("/home/pi/output .txt" j
ab" indicates ‘append binary’. If the file exists, output
WatfPi [ Chapter Ten )
66
will then be added after the existing file contents; if the file doesn
exist, it will be created and output will start at the beginning.
Sometimes when accessing a file, we don’t necessarily want to start
at the beginning. The f seek function can be used to reposition the file
pointer within the file.
[CHECK
THE LIBRARY]
#include <stdio.h>
void main (void)
{
FILE *fp;
int value;
fp = fopen ("/home/pi/input.txt", "rb");
if (fp)
{
fseek (fp, 10, SEEK_CUR);
while (1)
{
value = fgetc (fp);
if (value == EOF) break;
else printf ("%c", value);
}
fclose (fp);
The C library
offers a wide
range of
functions for
reading and
writing data
from and to
files; we’ve
just looked at
some of them.
If you need to
access a file,
have a look at
some of the
other library
functions,
like f read,
fwrite,
fscanf, fputs
and fgets,
to see if they
are more
appropriate
than the basic
fputc and
fgetc we've
used here.
}
The line fseek (fp, 10, SEEK_CUR) moves the file pointer to
bytes ahead of the current position, so this program will print all but
the first ten characters in the file. The first argument to fseek is the
file pointer; the second is the offset by which the pointer is to move.
This can be positive or negative; so fseek (fp, -5, SEEK_CUR)
moves the pointer 5 bytes back from the current position.
The third argument to fseek allows you to choose a position relative
to the start of the file (SEEK_SET) or the end of the file (SEEK_END) rather
than the current position. So fseek (fp, 12, SEEK_SET) positions the
pointer 12 bytes ahead of the start of the file, while fseek (fp, -17,
SEEK_END) positions it 17 bytes back from the end of the file.
[ File input and output ] !W 00*1
67
ESSENTIALS
CHAPTER ELEVEN
MORE ABOUT
TYPES AND
VARIABLES
Global variables, type definitions,
enumerations, and structures
[ LEARN TO CODE WITH C ]
■ pi@>raspberrypi: ~
Ostruct.c - /home/pi -
struct.c - /home/pi - Geany
File Edit Search View Document Project Build Tools Help
| * I *5 * v 4 | ® I
> ' struct.c > j
The typedef keyword allows you
to create your own new data types
4 a 1
<J Symbols
(p Functions
CP add [9]
main [14]
Structs
Fl $7 anon_struct_0
*0 invall [4]
Inval2 [51
“0 outval [6J
- $> Typedefs/Enum^
§> MY_DATA [71
Si
iclude <stdio.h>
Gtypcdcf struct { »■■■
int invall;
int inval 2 ;
int outval;
> HY_DATA;
void add (HY DATA »d)
a(
d->outval = d->invall + d->inval2;
void main (void)
HY_DATA data;
data.invall = 5;
data.inval2 = 7;
add (&data);
printf ("The sum of Vi and %d is ^d\n"
File Edit Tabs Help
piQraspberrypi:- $ gcc -0 myprog struct.<
piQraspberrypi: $ ./myprog
The sum of 5 and 7 is 12
piQraspberrypi:~ $ |
data.invall, data.inval2, data.outval);
| 10 : 11 : 01 ; RIe /home/pi/examples.c openedO).
Status 10 : 11 : 26 : New file "untitled’ opened.
10:11:41: File/home/pi/struct.c saved.
Compiler | 10:11:43: Rle/home/pi/examples.c closed.
^ /home/pi/input.txt closed,
A data structure allows you to
combine multiple items of data
in the same type
INS TAB mode: Unix (LF) encoding: U 1 F-8 filetype:C scope: main
NAMES]
While it’s
perfectly valid
to give a local
variable the
same name
as a global
variable in the
same program,
don't do it! If
you have a
global and a
local with the
same name,
the local
version is used
in the function
in which it
is declared,
and the
global is used
everywhere
else - this
can result in
unexpected
behaviour.
O n this chapter, we’re going to look at some of the more
advanced topics around the use of variables and types,
including the difference between local and global variables,
defining new types, and the use of enumerations and data structures.
When we’ve used variables in the examples in this book, we’ve
always put them inside function definitions. These are therefore local
variables-, that is, variables which are local to those functions and have
no meaning outside the function.
Global variables
C also allows global variables; that is, variables which are defined outside
all functions. These have global scope, which means they can be read and
written from any function within the program. Let’s look at an example:
#include <stdio.h>
int result;
void add (int a, int b)
{
[ More About Types and Variables ]
69
result = a + b;
>
void main (void)
{
add (3, 4);
printf ("The result is %d\n"j result);
}
In this example, the variable result is global. It can therefore be
read or written within both the add function and the main function;
as you can see, we write a value to it in add and read it back in main,
and so we don’t need to return a value from add.
In some ways, this looks easier than passing values about all over
the place, surely? So why not just do this all the time? The answer is
memory. Local variables in functions are temporarily allocated space
while the function is running, and the memory is freed up as soon as
the function ends. But global variables are allocated space when the
program starts, and that space isn’t freed until the program ends; if you
allocate enough of them, you can run out of memory on some systems.
There’s a better way of making a lot of data available to every
function, which we will come to a bit later on...
Type definitions
In a previous chapter, we looked at the range of variable types in
C: char, int, float, and so on. C also allows you to define your own
types, with what is known as a typedef. A typedef is a line of the format
typedef <existing type> <new name>, usually put at the start of a
program. For example:
typedef unsigned char BYTE;
This defines a new type called BYTE, which is another name for an
unsigned char. (Note that by convention, user-defined types are
usually given names in capital letters. It’s not compulsory, but it does
help to distinguish them from variables when reading the code.)
When we say this defines a new type, what it really does is to create
an alias to an existing type. This seems a bit pointless, but it can
1
[ LEARN TO CODE WITH C ]
help in two ways. First, it can make it more obvious what your code
is doing if you make the type names specific to your program’s data.
Second, by defining specific types, you can get the compiler to warn
you if you use the wrong type for a function argument or variable.
There are a couple of specific cases where typedefs are particularly
useful: these are enumerated types and data structures.
Enumerated types
Often, there’s a use for a variable which can only take one of a few
possible values. C provides a type called enum for this purpose, which
defines an integer with a fixed set of named values. Here’s an example:
#include <stdio.h>
typedef enum {
false,
true
} BOOLEAN;
void main (void)
{
BOOLEAN b_varj
[ NUMBERED
ENUMS ]
When you
create an
enum, the
compiler
assigns a
numeric value
to each of
the possible
values. By
default, it
numbers the
first in the
list as o and
counts up from
there. You can
override this
by putting an
equals sign
after each
named value
and setting it
to the value
you want.
b_var = false;
if (b_var == true)
{
printf ("TRUE\n");
}
else
{
printf ("FALSE\n");
}
>
mi
As you can see, the named values of the enumerated type are used
instead of numbers for assignments and comparisons. This can make
code a lot easier to understand, and is a very good way of preventing
errors, as an enumerated variable can only ever be set to a valid value.
[ More About Types and Variables ) 1Ua0 > i
71
ESSENTIALS
[ NEW
TYPES INSIDE
STRUCTURES ]
A structure
can contain
other new
types (plain
types, enums
or indeed other
structures);
just make sure
the typedefs
for them occur
before the
typedef of
the structure
in which
you want to
include them. .
Structures
The other really useful thing you can do with a typedef is to use it to
define a data structure. This is a collection of individual variables which
are grouped together, allowing you to pass the structure between
functions rather than the individual variables.
Here’s an example:
#include <stdio.h>
typedef struct {
int invall;
int inval2;
int outval;
} MY_DATA;
void add (MY_DATA *d)
{
d->outval = d->invall + d->inval2;
}
void main (void)
{
MY_DATA data;
data.invall = 5;
data.inval2 = 7;
add (&data);
printf ("The sum of %d and %d is %d\n"
data.inval2, data.outval);
data.invall.
So here we use a typedef to create a data type called MY_DATA The
definition of the structure consists of the keyword struct with a list of
variables enclosed by curly brackets; in this case, the structure consists
of three integer variables.
In the main function, we declare an instance of the structure as a
variable called data of type MY_DATA. We then access the individual
72
WatfPi [ Chapter Eleven ]
[ LEARN TO CODE WITH C ]
Below An instance of the MYDATA structure is used
to pass the three integers to the add function
pi@raspberrypi: <
File Edit Tabs Help
pi@raspberrypi:~ $ ./myprog
The sum of 5 and 7 is 12
pi@raspberrypi:~ $ |
elements of the structure by giving the name of the structure
variable (data), a full stop (.), and the name of the specific element.
So the line data. invall = 5 sets the value of the element invall
of data to 5, and so on.
The function add takes a pointer to a MY_DATA structure as its
only argument; as ever, a function cannot change the values of its
arguments, but can change values pointed to by its arguments,
so we pass a pointer rather than the structure itself.
To access the elements of a structure from a pointer to it, we replace
the full stop with an arrow made up of a minus sign and a greater than
sign (->). So the add function reads the values of invall and inval2 in
the structure pointed to by d, and then writes the result back to outval
in the same structure; the main function then prints the result from
the structure.
Structures are very useful if you need to pass a lot of data around
between functions; they can be a lot more memory-efficient than
having large numbers of global variables, as you only need to create
the structure as and when you need it, rather than taking up memory
all the time.
[.VS ->]
When
accessing a
structure’s
elements,
make sure you
use the right
symbol. A. is
used if your
variable is an
instance of the
structure itself;
a -> is used if
your variable
is a pointer to
an instance of
the structure.
Using the
wrong one will
usually give
an error from
the compiler.
[ More About Types and Variables ] MckfPi
73
ESSENTIALS
CHAPTER TWELVE
HEADER FILES
AND THE
PREPROCESSOR
Splitting code up into multiple files
[ LEARN TO CODE WITH C ]
© a ■ B pi@raspberrypi: ~ | jcircle.c - /home/pi - G.
Wastebasket
<3P circle.c - /home/pi - Geany
File Edit Search View Document Project Build Tools Help
C? v ^ v © X ! *5 % v | ®
<1 Symbols
- Functions
$ main [4]
Macros
3 > PI 12]
Every use of the symbol PI is
repLaced by the value in the
matching #define directive
'> circle.c X
finclude <stdio.h>
#define PI 3.14159
void main (void)
(=H
float rad = 3;
float circ = rad
float area = rad
2 * PI;
rad * PI;
printf ("The circumference of a circle radius Vf is Vf\n", rad, circ);
printf ("The area of a circle radius Vf is %f\n*« rad, area);
pi(8>raspberrypi>
File Edit Tabs Help
Qraspberrypi: - $ gcc -o myprog cir
piQraspberrypi: ' $ ./myprog
The circumference of a circle radius
The area of a circle radius 3.00000G
piQraspberrypi:- $ |
All preprocessor directives, like
#define, start with a # sign
Compiler 1 6:38:02: File/home/pi/circle.c saved.
O il the examples we’ve seen so far have put all the code for a
program in a single C file. But once programs get big, it makes
more sense to be able to split them up into separate files,
grouping similar functions together. To understand how this works,
we need to look in more detail at what the compiler actually does.
In all the examples so far, we’ve called gcc on a single source file and
it has created a single executable program. This hides the fact that gcc
actually does two things: first, it compiles your C source file into what’s
called an object file, and then it links the object file with all the library
functions to create the executable. This second step is performed by a
program called a linker, gcc actually does both jobs.
If you create a program with multiple source files, you just need to
include the names of all the source files in the call to gcc. It will then
create one object file for each source file, and then link all your object
files together to create the executable.
There’s one snag, though. If you’ve separated your code into separate
files (usually referred to as modules ), you’ll have some files which
make calls to functions in other files in order to work. These files don’t
find out about each other until the linker operates on them; the files
are compiled individually, and the compiler will complain if you use
functions in a file it doesn’t know about.
I
[ Header Files and the Preprocessor ]
75
[ KEEP NAMES
CONSISTENT]
While you
can call
header files
whatever you
like - there's no
magic about
the names - it's
good practice
to give the
header file for
the functions
in a particular
C file the same
name as the C
file itself, with
a . h extension
rather than .c.
This makes
it easier for
someone
reading your
code to find
the files where
We fix this using header files. These are files with the extension . h
which hold the declarations of functions defined in a module, so that
the compiler can be told about them when they’re used by another
module. We’ve already seen this many times; remember that line
#include <stdio.h> at the top of the examples? That is exactly this
process; it’s telling the compiler that functions declared in the system
header file stdio. h are used in this module.
Splitting code into multiple files
Let’s look at an example of how this works. Create three files, two with
the extension . c and one with the extension . h , as follows:
function.c
int add_vals (int a, int b } int c)
{
return a + b + c;
>
function.h
extern int add_vals (int a., int b, int c);
main.c
ttinclude <stdio.h>
#include "function.h"
void main (void)
{
printf ("The total is %d\n" J add_vals (1, 2, 3));
>
Put all three files in the same directory and run gcc, giving it the
names of both . c files - gcc -o myprog main.c function.c.
The resulting program will run the main function from main.c,
which calls the add_vals function from function, c.
A few things to note. First, inside the header file we declare the
76
IWatfPi [ Chapter Twelve ]
[ LEARN TO CODE WITH C ]
function with the word extern at the start of the declaration. This tells
the compiler that this function is to be found externally to the file, i.e.
in another C file.
Second, while we have always included stdio.h with its name
between <> signs, we include function, h in double quotes. The <>
signs tell the compiler to look for the file in the directory where the
system’s include files are stored; the "" signs indicate that the file
is local and is in the same directory as the . c files you’re building. If
you’re creating your own header files, always use double quotes around
the name when including them.
The preprocessor
So what does ffinclude actually do? It’s an instruction to the preprocessor,
which is the first stage of compiling; it substitutes text within source files
before passing them to the compiler itself. The preprocessor is controlled with
what are called directives; these are easy to spot, as they all start with a # sign.
The #include directive instructs the preprocessor to replace the line with
the file which it’s including. So in our example above, the line ttinclude
"f unction, h" in the .c file gets replaced with the contents of the file
function, hj meaning that what’s passed to the compiler looks like:
#include <stdio.h>
extern int add_vals (int a, int b, int c);
void main (void)
{
printf ("The total is %d\n"j add_vals (1, 2 , 3));
>
Left The add_vals
function is called
from the main
function - the
linker connects
the caLl from
main.c to the
function definition
in function.c
[ MAKEFILES ]
As you can
imagine, if you
have a project
with tens or
hundreds of
C files, typing
all their names
in the call to
gcc every time
would be a
bit tedious!
Large projects
are built with
a tool called
"make”, which
stores build
instructions in
a "Makefile”.
Makefiles
are outside
the scope of
this book, but
there's lots of
information i
about
them online.
[ Header Files and the Preprocessor ] McifJPi
77
[ DON'T
INCLUDE
.C FILES ]
#define
Another useful directive is ftdefine, which can be used to define
constant values. Look at this example:
ffinclude <stdio.h>
#define PI B. 14159
void main (void)
{
float rad = 3;
float circ = rad * 2 * PI;
float area = rad * rad * PI;
printf ("The circumference of a circle radius %f is %f\n",
rad, circ);
printf ("The area of a circle radius %f is %f\n", rad, area);
>
The directive #define is used to set the value of pi. The important
thing to remember is that PI isn’t a variable; it’s text that will be
substituted by the preprocessor. The #define line tells the preprocessor
to go through the file and replace every instance of the symbol PI with
the digits 3.14159 before passing it to the compiler. So a line which
does something like PI = 5; will cause an error; the compiler
will see the meaningless statement 3.14159 = 5;.
Why is this useful? Why not just declare a float variable called
PI and set it to 3.14159? A floating-point variable requires
allocating memory in which to store it; using #define saves
that memory, which is useful if memory is limited.
You can also ffdefine functions:
ffinclude <stdio.h>
ffdefine ADD(a,b) (a+b)
void main (void)
{
printf ("The sum of %d and %d is %d\n", 5, 2, ADD(5,2));
printf ("The sum of %d and %d is %d\n", 3, 7, ADD(3,7));
}
78
MatfPl [ Chapter Twelve ]
[ LEARN TO CODE WITH C ]
Again, this does a text substitution; whenever ADD(a, b) appears
in the code, it’s replaced by (a+b), with the values of a and b replaced^
by the arguments to ADD.
The preprocessor can also evaluate conditions with the #if directive:
#include <stdio.h>
void main (void)
{
#if 0
printf ("Some code\n");
#else
printf ("Some other code\n");
#endif
>
[#DEFINES
FOR TEXT]
If you use
ftdefine for
text strings,
they should
be enclosed in
double quotes,
otherwise
the replaced
text will end
at the first
space. So use
ttdefine MY_
TEXT "This
is some
text to
replace."
The double
With a o after the #if, the code between the #if and the #else
doesn’t get called, but the code between the #else and the #endif
does. If you change the value after the #if to a 1, the code between the
#if and the #else does get called, but the code between the #else and
the #endif doesn’t. This is a really useful trick to temporarily remove
or replace a piece of code when you’re debugging.
quotes are
included in the
replacement,
so you can
then just call
printf (MY_
TEXT);
Left The most
common useofffif
is for temporarily
removing code
-just wrap it
between an #if 0
and an #endif. The
#else is optional,
but sometimes you
want to substitute
the code you've
removed with
different code.
[ Header Files and the Preprocessor ] !W citfPi
79
NEXT
STEPS
his book is intended as an introduction
i to the C programming language
n for beginners. It’s not a complete
description of every detail of C; such a book
would be at least twice the length of this one,
and would cover a lot of material that even
experienced programmers don’t use on a
regular basis.
The definitive reference on C is the book
“The C Programming Language”, by Kernighan
and Ritchie, now in its second edition. If you’ve
followed and understood all the material here,
and want to know about the more advanced
aspects of C, it’s a good place to look next.
I
ESSENTIALS
CHAPTER THIRTEEN
QUICK
REFERENCE
ESSENTIALS
. | CONTROL
STRUCTURES
If
if (<test>)
<code executed if test is true>
If-else
if (<test>)
<code executed if test is true>
else
<code executed if test is false>
Multiple if-else
if (<testl>)
<code executed if testl is true>
else if (<test2>)
<code executed if testl is false and test2 is true>
else
<code executed if testl is false and test2 is false>
82
IMat/Pi [ Chapter Thirteen ]
[ LEARN TO CODE WITH C ]
Switch ■
switch (<variable>)
{
case <testvall> : <code executed if variable is testvall>
break;
case <testval2> : <code executed if variable is testval2>
break;
default :
<code executed if variable is neither
testvall nor testval2>
break;
>
Switch with fall-through
switch (<variable>)
{
case <testvall> : <code executed if variable is testvall>
case <testval2>
: ccode executed if variable is either testvall
or testval2>
break;
default :
<code executed if variable is neither
testvall nor testval2>
break;
>
In all loops, the keyword break can be used to exit the loop and resume
execution immediately after the loop.
In all loops, the keyword continue can be used to skip code remaining in the
body of the loop and resume execution at the next iteration of the loop test.
[ Quick Reference 1 1Ua0 > i
While
1
while (<test>)
<code executed repeatedly while test is true>
Do-while
do
<code executed once and then repeatedly while test is true>
while (<test>)
For
for (cinitial conditions <increment>; ctermination condition>)
<code executed repeatedly until termination condition is true>
84
MatfPi [ Chapter Thirteen ]
VARIABLE TYPES
Name
Description
Size
(bytes)
char
Single alphanumeric character
1
signed char
Signed 8-bit integer (-128 - 127)
1
unsigned char
Unsigned 8-bit integer (0 - 255)
1
short, signed short
Signed 16-bit integer (-32768 - 32767)
2
unsigned short
Unsigned 16-bit integer (0 - 65335)
2
int, signed int
Signed 32-bit integer (-2147483648
-2147483647)
4
unsigned int
Unsigned 32-bit integer
(0 - 4294967295)
4
long, signed long
Signed 32-bit integer (-2147483648 -
2147483647)
4
unsigned long
Unsigned 32-bit integer (0 - 4294967295)
4
float
Floating-point value (+/- 3.402823 x 10 3S )
4
double
Double-precision floating-point value
(+/- io 30S )
8
III
Depending on platform, int can be either a short int (16 bits) or a
long int (32 bits); on Raspbian, as per the table above, int is a long
(32 bit) integer value.
[ Quick Reference 1 !W ac/Pi
85
The
1 FORMAT
SPECIFIERS
Specifier
Format / type
%C
Alphanumeric character - char
%d
Signed decimal value - int
%ld
Signed decimal value - long int
%u
Unsigned decimal value - int
%lu
Unsigned decimal value - long int
%o
Octal value - int
%lo
Octal value - long int
X
X
Hexadecimal value - int*
%lx, %1X
Hexadecimal value - long int*
%f
Floating-point value - float
%e
Exponential value - float
%s
Text string - char pointer
* %x displays a value as hexadecimal with lower-case letters a through
f; %X displays it with upper-case letters A through F.
I
[ LEARN TO CODE WITH C ]
The width (or minimum number of characters printed) can be set
by inserting a number between the % and the letter; this will pad a
value shorter than this with spaces at the start. To pad with spaces at
the end, insert a - between the % and the number. To pad with leading
zeroes, insert a 0 between the % and the number.
For example, to print an integer variable with the value 42, using
the format specifier "%5d" will print 42 with three spaces before it.
The format specifier "%-5d" will print 42 with three spaces after it.
The format specifier "%05d" will print it as 00042.
The number of decimal places shown for a floating-point or
exponential value can be set by inserting a decimal point followed
by a number between the % and the letter; this can be combined
with a width by putting the width before the decimal point.
For example, to print a floating-point variable with the value
76.54321, using the format specifier "%.2f " will print it as 76.54.
The format specifier "%08.2f" will print it as 00076.54. (Note that
the decimal point takes up one character of the specified width.)
OPERATORS
The operators in the table below produce a result which can be assigned to
another variable, i.e. c = a + b, but do not affect the values of a or b.
Symbol
Function
a + b
Addition
a - b
Subtraction
a * b
Multiplication
a / b
Division
a % b
Modulo (calculate remainder)
[ Quick Reference 1 1Ua0 > i
ESSENTIALS
a & b
Bitwise AND
a | b
Bitwise OR
a A b
Bitwise XOR
a << b
Bit shift left
a >> b
Bit shift right
~a
Bitwise l’s complement
! a
Logical NOT
The operators in the table below modify the value of a directly.
Symbol Function
Symbol
Function
a++
Increment a by one*
a- -
Decrement a by one*
++a
Increment a by one*
- -aa
Decrement a by one*
a += b
Increment a by b
a -= b
Decrement a by b
a *= b
Multiply a by b
a /= b
Divide a by b
a %= b
a = remainder of a / b
88
MatjPi [ Chapter Thirteen ]
[ LEARN TO CODE WITH C ]
a &= b
Bitwise AND a with b
a | = b
Bitwise OR a with b
a A = b
Bitwise XOR a with b
a <<= b
Bit shift a left by b
a >>= b
Bit shift a right by b
*The difference between a++ and ++a is if they are used in a test, such
as if (a++), a++ tests the value and then increments it, while ++a
increments the value first and then tests the incremented value.
The operators in the table below are used for comparisons in tests.
Symbol
Function
==
Is equal to
! =
Is not equal to
>
Is greater than
<
Is less than
>=
Is greater than or equal to
< =
Is less than or equal to
III
[ Quick Reference ] MckfPi
89
ESSENTIALS <9
LEARN I CODE I MAKE
AVAILABLE NOW:
> CONQUER THE COMMAND LINE
> EXPERIMENT WITH SENSE HAT
> MAKE GAMES WITH PYTHON
> CODE MUSIC WITH SONIC PI
> LEARN TO CODE WITH SCRATCH
> HACK & MAKE IN MINECRAFT
> ELECTRONICS WITH GPIO ZERO
> LEARN TO CODE WITH C
From the makers of the
official Raspberry Pi magazine
ESSENTIALS
OUT NOW
IN PRINT
ONLY £ 3.99
ESSENTIALS
raspberrypi.org/magpi
Live Music Archive
Librivox Free Audio
Metropolitan Museum
Cleveland Museum of Art
Internet Arcade
Console Living Room
Open Library
American Libraries
TV News
Understanding 9/11